Ultimate Base Running Primer

Base running linear weights or base running runs, or Ultimate Base Running (UBR), is similar to the outfield arm portion of UZR. Whatever credit (positive or negative) is given to an outfielder based on a runner hold, advance, or kill on a batted ball is also given in reverse to the runner (or runners). There are some plays that a runner is given credit (again plus or minus) for that do not involve an outfielder, such as being safe or out going from first to second on a ground ball to the infield, or advancing, remaining, or being thrown out going from second to third on a ground ball to SS or 3B.

Runs are awarded to base runners in the same way they are rewarded to outfielders on “arm” plays. The average run value in terms of the base/out state is subtracted from the actual run value (also in terms of the resultant base/out state) on a particular play where a base runner is involved. The result of the subtraction is the run value awarded to the base runner on that play.

If you didn’t understand that, a simple example should explain it clearly:

Let’s say that there is a runner on second and one out. A ground ball is hit to the SS. Let’s say that on the average, in that same situation, the runner advances safely to third and the batter is thrown out 20% of the time, he stays put 70% of the time, he gets thrown out at 3rd 5% and beats a throw to third 5% of the time (batter safe on a FC). And let’s say that average base/out run expectancy (RE) of all those results, weighted by their frequency of occurrence, is .25 runs (all the numbers are made up). If the runner advances and the batter is thrown out, and the resultant RE is .5 runs, then the runner gets credit for .25 runs (.5 minus .25). If he stays put, and the average RE of a runner on second and 2 outs is .23 runs, then gets “credit” (he gets docked) for -.02 runs (.23 minus .25). So basically a runner gets credit for the resultant run value of what he does minus the average weighted resultant run value of all base runners in that situation.

Here are most of the situations where a base runner gets some kind of positive or negative credit. Obviously more than one base runner can get credit on any particular play.

1) On a hit, advancing an extra base, not advancing an extra base, or getting thrown out trying to advance an extra base, as long as no other base runner is blocking an advance.

2) A batter getting thrown out trying to advance an extra base on a hit (if he successfully does, we don’t know it, as he is simply awarded a double, for example, on a usual single where he advances an extra base).

3) On a hit, the batter advancing, not advancing, or getting thrown out when a runner is safe or out advancing an extra base.

4) Trailing runners advancing, not advancing or getting thrown out when a leading runner is safe or out trying to advance an extra base on a hit or an out. This is basically lumped together with #1 above.

5) Runners trying to advance on fly ball outs – i.e. tagging up.

6) As mentioned above, on ground balls to the infield, runners on first staying out of the force or DP at second base, whether the batter is out or is safe on a FC.

7) Also as mentioned above, a runner on second advancing or not (or getting thrown out) on a ground ball hit to SS or 3B.

Runners on third base advancing, not advancing, or getting thrown out at home on a ground ball are not considered (on air balls they are). Runner advances or outs on WP or PB are not considered either.

All of these situations are considered an “opportunity” for the base runner or the batter (except when a batter gets a hit, he is not awarded an opportunity unless a leading runner tries to advance an extra base and the batter has an opportunity to advance on the throw).

As with UZR, a player’s “games” are not his actual games played. They are his opportunities divided by the league-average opportunities per game. No adjustments are made for how often a player typically gets on base (or how often subsequent batters put the ball in play or get hits), such that a player with a high OBP will likely have more “games” than actual games played and a player with a low OBP will likely have fewer “games” than actual games played.

On batted balls to the outfield, only whether it was hit to LF, CF, or RF is considered, not the depth or actual vector or zone within the three outfield positions.

Future versions of UBR will likely adjust “games” for a player’s OBP and will also likely include more outfield location zones. These upgrades are not likely to significantly change the numbers.



Print This Post





Sort by:   newest | oldest | most voted
m123
Guest
m123
5 years 3 days ago

Dustin Pedroia tripping over himself while running to third count?

Jon
Guest
Jon
5 years 3 days ago

Nothing Pedroia does is allowed to effect him negatively

it just adds to his grit

Jim
Guest
Jim
4 years 11 months ago

Are we seriously still doing this meme with Pedroia? He’s a legit superstar for god’s sake, not David Eckstein.

Michael
Guest
Michael
5 years 3 days ago

For your second point, couldn’t you use something similar to UZR (almost the inverse actually) to assign a base running value for taking the extra base on hits? As in, if the batter hits the ball to location X, there is an average number of bases the batter will advance when the ball is hit to that location. The difference between expected and the outcome should be their base running value. Maybe I’m being too simplistic about it, and there might be a very good reason why this would not work, but my small brain is having trouble figuring out why it wouldn’t at the moment.

chuckb
Guest
chuckb
5 years 3 days ago

I agree with this. There’s got to be a way to account for this, right?

I was also wondering if it was possible to deduct bases for double plays batted into when the average player would have beaten out the throw to first. In other words, shouldn’t a player lose credit if he’s unable to beat out a double play ball that the average batter would have beaten out.

MGL
Guest
MGL
5 years 3 days ago

Yes, that is in the plan for future versions of UBR. Maybe not so many buckets, since sample size is definitely going to be an issue.

gradygradychase
Member
gradygradychase
5 years 3 days ago

MGL is always soooooo awesome that he is now attempting to estimate a player’s clubhouse chemistry.

The question is, when will it appear in Our FanGraphs?

MBH
Guest
MBH
5 years 3 days ago

Is it possible to look at baserunning like they are now looking at putting on the PGA tour. Once someone is faced with a putt, they compare the number of strokes it takes him to get down to the tour average from that position…he’s better then average, equal to or worse. If we look at all the situations where a baserunner is on first with no outs and a ground ball single to right…what are avg number of bases advanced, then compare baserunner A’s chances to all those avg chances…

MGL
Guest
MGL
5 years 3 days ago

As far as I can tell from your post, that is exactly what UBR is doing now. It just does it one play at a time, which will yield exactly the same results as what you are proposing.

Ted
Guest
Ted
5 years 3 days ago

Very cool. Is it possible to turn this into a rate state, like UZR/150?

Would probably need to be tied to hits/times on base rather than PA though, to avoid penalizing good baserunners who never get on base.

MGL
Guest
MGL
5 years 3 days ago

Not sure if they (FG) are posting “base running games”. If they are, then it is done. If not, you can use their actual games played to figure the rate…

whatzitmather
Member
whatzitmather
5 years 3 days ago

Braves rate 29th in BP’s EQBRR, but are midpack in UBR. Is this explained mostly by the fact that SB/CS are omitted from UBR, or by the fact that Chipper Jones is a pretty substantial outlier when you compare UBR to EQBRR (as show by Chris St. John here: http://sbn.to/jVVW2p ). I’m not sure I understand how there could be such a big difference in overall rankings unless the SB/CS are weighted heavily in EQBRR? Could someone help me out?

MGL
Guest
MGL
5 years 3 days ago

Don’t know. I’m sure there are a few minor differences between the two systems. Minor differences will always show up as a major difference for a few players…

CSJ
Member
5 years 2 days ago

There are 5 additive parts of EqBRR: EqGAR (Ground Advancement), EqSBR (Stolen Base), EqAAR (Air Advancement), EqHAR (Hit Advancement) and EqOAR (Other Advancement).

On average, these account for this percentage of EqBRR:
GAR: 52%
SBR: 4.7%
AAR: 7.5%
HAR: 35%
OAR: 0.27%

longgandhi
Guest
longgandhi
5 years 3 days ago

So if I understand this correctly, you are going to give credit to the baserunner when he advances because the outfielder throws to the wrong base?

MGL
Guest
MGL
5 years 3 days ago

Well, if you have some way of telling me what base an OF throws the ball to (I don’t), I’ll be glad to incorporate that into the metric. Why don’t you email me or post it on FG every time that happens and then compile it at the end of the year and send me the data? While you’re at it, maybe you can do my taxes for me.

What a moronic comment…

MikeS
Guest
MikeS
5 years 3 days ago

I disagree that the question is moronic. I was at a game two weeks ago, men on 2nd and 3rd, none out. Hitter hits a sinking line drive to left, LF charges, makes a diving catch, gets up and airmails the cut off man. While the man on 3rd scores standing up, the other runner takes 3rd when he never would have if the ouytfielder threw to third which was the proper play since he had no shot at home. The question is if the runner gets credit for the fielder screwing up. As for knowing which base was thrown to, I assumed this was being done by stringers since you talk about whether a runner would have taken a base on average. only way to know that is to see the play. Did the SS go hard to his left to field a dribbler up the middle or was it a one hopper hit up the middle? Was it a hit and run?

I am afraid to ask why runners on 3rd get no credit for coming home on a ground ball or whether the system incorporates non batted ball plays like PO, SB or CS.

We have a right to judge the accuracy of metrics used and the only way we can do that is if we understand the methods. Any good researcher knows that the first section you read of any paper is the methods so you can spot fundamental errors. besides, belittleing an audience that is trying to understand your post only chases people away.

Welch
Guest
Welch
1 year 1 month ago

Example: classic Daryl Strawberry play. Runner on 2B. Single to RF. Strawberry charges, throws a perfect line to catcher. Gary Garter takes it belt-high, but the runner has scored standing and the batter took second. Impressive throw but bad baseball.

The hitter probably should get some credit for taking second if he saw that Strawberry had ignored the cutoff man. Or maybe the 1B coach waved the hitter to second? Or maybe the hitter intended to go for second to draw a throw and allow the runner to score more easily? It’s hard to decide.

J-Doug
Guest
5 years 3 days ago

It’s moronic to ask if your metric properly accounts for defense?

chuckb
Guest
chuckb
5 years 3 days ago

longgandhi’s comment reads snarky and sarcastic. Perhaps that’s not how he intended it, but if MGL read it sarcastically, he’s not alone. Reading it again, I still think it was meant sarcastically and, if I was MGL, wouldn’t appreciate it very much either.

Sensitive RAB Guy
Guest
3 years 5 months ago

i want to fight you

T
Guest
T
5 years 3 days ago

How big a sample does this need to have before it can be conclusive?

MGL
Guest
MGL
5 years 3 days ago

A million games? An infinite number of games? Define “conclusive?”

T
Guest
T
5 years 3 days ago

Draw judgments upon it that are indicative of true talent. No need to get snarky.

T
Guest
T
5 years 3 days ago

*from it, not upon it.

MGL
Guest
MGL
5 years 3 days ago

You can draw judgments at any point – seriously. How much certainty there is in those judgments is another story. I’ll look at the year to year correlations and let you know what the 50% regression point is…

sleepingcobra
Guest
sleepingcobra
5 years 3 days ago

Is MGL having a bad day or what? What’s with all his pissy attacks in the comments section?

By the way, I love your new stat, can you please punch me in the face now, too?

Jim
Guest
Jim
5 years 3 days ago

I think he’s upset that people are asking honest questions as opposed to just dropping to their knees and thanking him breathlessly.

Anon
Guest
Anon
5 years 3 days ago

Having been on the book blog before, this isn’t remotely surprising.

xeifrank
Member
5 years 3 days ago

This is common for him. You should read one of his non sports (plus comments) posts over at his blog. Oh boy! Great baseball mind though.

Rodion
Guest
Rodion
5 years 2 days ago

It’s nothing new. I first discovered sabermetrics when I was 14 years old. I approached the subject ingenuously and with unbridled enthusiasm. I made the mistake of stumbling upon the fanhome – sabermetrics forums and posted a couple of questions and ideas there; after being insulted multiple times by MGL, I left the site and nearly lost my enthusiasm for the subject.

The discussions that are generated from new ideas and statistics are often just as valuable as the ideas themselves, and help contribute to a deeper understanding of the game. It’s a shame such a bright mind can’t appreciate that.

MGL
Guest
MGL
5 years 3 days ago

Looks like the year to year correlation for full time players is around .5, which means that one year of full time UZR requires around a 50% regression. Two years 33% regression, 3 years, 25%, etc.

I really don’t give a crap whether anyone likes any of these stats or not. Use them at your pleasure or displeasure.

When you spend more than 1 minute on a piddling post as opposed to hundreds of hours coming up with these metrics, then you can talk. It’s not rocket science but it does take time…

Yirmiyahu
Member
Yirmiyahu
5 years 3 days ago

Assume you meant UBR?

T
Guest
T
5 years 3 days ago

As far as I can tell, only one person in this entire thread is criticizing anything about this stat at all, and he was also clearly assuming you were using a database that indicated something that it obviously doesn’t indicate. Everyone else was asking how to use it.

MGL
Guest
MGL
5 years 3 days ago

If you’d rather I not answer any questions with or without snark, then don’t ask any. If not, take them as they come and filter them as you like…

MikeS
Guest
MikeS
5 years 3 days ago

Dude, calm down. You introduce a brand new stat and people want to know how it works, what it can do, what the limitations are, etc. These people are showing interest in your work, not criticizing it. They want to understand it and use it. I haven’t seen anybody here saying it’s a load of crap, we are just trying to learn more about what you have done. What’s wrong with that?

T
Guest
T
5 years 3 days ago

I’m going to assume you’re having a bad day, because your attitude here has been abhorrent.

Andre
Guest
Andre
5 years 3 days ago

Agreed.

Chris Long
Guest
Chris Long
5 years 3 days ago

I’m going to call my system Awesome Base Running.

Lee Panas
Guest
5 years 3 days ago

Great work MGL. I’m looking forward to tracking this stat. I’ve been using EqBRR a lot, but I find the FanGraphs database to be easier to use than the BP database. So, I’m glad that there is now going to be a baserunning stat here.

pft
Guest
pft
5 years 3 days ago

It occurs to me that players who play on teams that win or lose big more often league average may be less aggressive on the bases and be penalized to some extent, but it probably does not add up to much.

In 2010, the spread between the best runners and worst was 16.6 runs. This seems about right if SB are excluded

“Runners on third base advancing, not advancing, or getting thrown out at home on a ground ball are not considered (on air balls they are).”

Seems it should be considered, but then again, there are many cases where the defense concedes the run due to the game situation, so maybe not.

I also think players who stay out of GIDP due to their speed should be credited (eg Carl Crawford GIDP only 2 times in 2010, or a sickly 2% of GIDP opportunities. And players who hit into an extraordinary amount of DP due to their slow speed or high GB rate should be penalized.

Did a bit of a double take when I saw Ichiro was minus in 2010, but I suppose this might have something to do with the quality of the offense hitting behind him.

While I do like having a single number, it would be nice to see the components, 1st to home, 2nd to home, 1st to 3rd, thrown out, picked off, tagged up, etc.

Players who play for teams with awful 3B coaches are probably unfairly penalized. Bogar with the Red Sox is a good example, so I can’t help but wonder if the 2011 Red Sox are last in the AL in UBR is due to plays at home or 3B.

reillocity
Member
reillocity
5 years 2 days ago

For what it’s worth, I second the idea of incorporating run differential and inning into a future iteration of the stat. I’m sure that’s a whole lot more work, too.

MGL
Guest
MGL
5 years 3 days ago

Yup, missed a 4-foot putt on the 18th hole today for a 10-skin carry. Threw my putter across the water (tried to throw it IN the water) and then threw my bag into the woods. My son thought I was acting like a child. I told him that I ran out of my Lexapro for a couple of days.

Anyway, I always try and answer all questions as best as I can, either on the blog or emailed to me…

Kenny Ocker
Member
5 years 3 days ago

“What a moronic comment…” is no way to “answer all questions as best as I can.”

You could have just as easily said to him, “It is impossible to statistically determine where the outfielder is throwing the ball on each play, because box scores do not denote where the ball is thrown when there is no assist made. Because of that, the metric gives credit to runners who advanced on suboptimal throws.”

But you didn’t. You didn’t answer the question. Instead, you ripped the person for trying to understand what you had presented.

How can you expect to get respect for your work when you’re unwilling to give it to anyone else, especially those eager to understand what you’ve presented?

Chris Long
Guest
Chris Long
5 years 3 days ago

It’s Mitchel, he’s the sabermetric version of the old man who yells at you to get off his lawn and who keeps any balls or frisbees that fall into his yard. He calls it as he sees it, and he makes me look nice in comparison, so leave him alone. :)

Will you be swinging through San Diego this season, mgl?

Chip Buck
Guest
Chip Buck
5 years 3 days ago

Great stuff MGL as always. I’m glad to see there’s an alternative to EqBRR, which I use from time to time. It’s also nice to see Fangraphs will be incorporating it into their version of WAR. What are your thoughts about adjusting for ballparks that can have a large affect on baserunning trends? For instance, balls hit to Fenway’s LF tend to minimize the number of times runners can take an extra base.

By the way…while I’m sure I’m in the minority, I rather enjoy the snarky responses. At the very least it’s been entertaining.

Kel
Guest
Kel
5 years 2 days ago

Pablo Sandoval was leading the league in UBR until this happened:

PaulScarfo
Guest
PaulScarfo
5 years 2 days ago

Mitch, do you ever work with statisticians when creating these metrics?

MGL
Guest
MGL
5 years 2 days ago

Chris, I probably won’t. I’m in NY for the summer. Next time I’m in SD, I’ll give you a holler.

“How can you expect to get respect for your work when you’re unwilling to give it to anyone else, especially those eager to understand what you’ve presented?”

Not looking for respect for baseball work.

That person who wrote about throwing to the wrong base, was not asking a question. They were being sarcastic as far as I can tell. In any case, I did answer his question quite explicitly. I clearly stated that I don’t have that information. As far as my tone, that is his problem, not mine.
t

longgandhi
Guest
longgandhi
5 years 2 days ago

Actually, I wasn’t being sarcastic. I was a stringer for MLB.com for 5 years and I was noting a misplay I would see with some frequency, and the way you described it, your metric appeared to give credit to the wrong player. I have no idea what data is available to you but given the preponderance of video, it doesn’t seem far fetched that someone could provide that data if requested. Through my contacts with two teams and BIS I know that teams request unusual data occasionally and your profile states that you worked for a team. If I made a mistake it was in thinking you were privy to such data. Telling me to compile and send you that data (for free, I’m assuming) and then suggesting I do your taxes contradicts your assertion that the tone problem is solely mine. No worries – no harm, no foul.

PaulScarfo
Guest
PaulScarfo
5 years 2 days ago

There’s no problem. Mitchel doesn’t care what you think.

MGL
Guest
MGL
5 years 2 days ago

It’s all good. I did think you were being sarcastic. No problem.

Obviously errors in the data (or incomplete information) are not going to make much difference overall. Other than the randomness involved in any sample metric, this is one of the purer ones. The fact that there are not that many opportunities per season yet the y-t-y correlation is at least .5 suggests that there is not much noise in this metric, compared to other metrics.

glassSheets
Guest
glassSheets
5 years 2 days ago

If a can of corn is lost in the lights, the hitter gets a double and his WAR goes up. If the runner advances on a dumb throw, the runner gets credit and his WAR goes up. It happens with most stats, not just the new ones. Fielders make bad throws, Aubrey Huff plays the outfield, balls take bad hops or hit the bases, the wind blows out at Wrigley when the Rockies are there and the wind blows in at Wrigley when the Giants are there. The signal is a larger magnitude than the noise in this case.

Payton Kienle
Guest
Payton Kienle
9 months 7 days ago

Is there an actual formula for this?

wpDiscuz